Re: Stress placement systems

On 9/20/06, R A Brown <ray@...> wrote:
> Dirk Elzinga wrote:
> > For what it's worth, I just looked in Hayes 1995, and of Ancient Greek
> > he says (p 181): "Ancient Greek [has Fijian-like stress] in the
>
> Thanks - it doesn't change my opinion    :)
I didn't think it would. But it does show that the database
maintainers probably paid undue attention to Hayes (1995) and didn't
do any checking of their own.

> > According to Hayes, in Modern Greek "Main stress is lexically
> > determined; limited to one of the last three syllables." (p 204)
>
> Yep - it is limited to one of the last three syllables. I think it would
> be more accurate to say it is lexically & morphologically determined.
> One thing is certain: it is not in any way conditioned by syllabic
> weight. Clearly a Syllable Priority Code is not going to work for modern
> Greek.
>
> My own feeling is that using a SPC to describe every language with word
> stress is not going to work.
I agree. I would actually want to *exclude* all systems that are
lexical and/or morphological.

It is perhaps significant that the database
> does not include modern Greek.
But Russian is *included*, and it is presumably a stress system which
is lexically and morphologically determined, just as you claim for
modern Greek. Clearly more homework needs to be done on the part of
the database maintainers.

If it is really going to be a Stress
> System Database, then it seems to me that it will have to use a code
> system that includes languages where syllabic weight is not relevant to
> stress. IMO at present it does not use a systematic way to describe all
> stress systems.
As I said above, I would want to exclude all non-phonological stress
systems. The remaining database, limited as it is, would still be a
useful resource, if sufficient attention were paid to accuracy and
fact-checking.

> Ray
Dirk

Reply